Boris Altshuler - Lancaster University · Introduction to Anderson Localization 2. Phononless...

The 4th Windsor Summer School on Condensed Matter Theory Quantum Transport and Dynamics in Nanostructures

Great Park, Windsor, UK, August 9 - 21, 2010

Many-Body Localization

Boris AltshulerPhysics Department, Columbia University

Outline:1. Introduction to Anderson Localization

2. Phononless conductivity

3. Localization beyond real space

4. Spectral Statistics and Localization

5. Many – Body Localization

6. Disordered bosons in 1D

7. Metal - Insulator transition in electronic systems

1.Introduction

>50 years of Anderson Localization

One quantum particle

Random potential (e.g., impurities) Elastic scattering

Einstein (1905):Random walk

always diffusion

Dtr =2

diffusion constant

Anderson(1958):For quantum

particles

not always!

constr t → ∞→2

It might be that

as long as the system has no memory

Quantum interference memory

2 0Dtρ ρ∂

− ∇ =∂

2 dne Dd

σ ν νµ

Density of statesConductivity

Einstein Relation (1905)

Diffusion Constant

2 dne Dd

σ ν νµ

Density of statesConductivity

Einstein Relation (1905)

Diffusion Constant

No diffusion – no conductivity

Metal – insulator transition

Localized states – insulatorExtended states - metal

Localization of single-electron wave-functions:

extended

localized

Disorder

Conductanceextended

localizedexp

extended

localized

I was cited for work both. in the field of magnetism and in that of disordered systems, and I would like to describe here one development in each held which was specifically mentioned in that citation. The two theories I will discuss differed sharply in some ways. The theory of local moments in metals was, in a sense, easy: it was the condensation into a simple mathematical model of ideas which. were very much in the air at the time, and it had rapid and permanent acceptance because of its timeliness and its relative simplicity. What mathematical difficulty it contained has been almost fully- cleared up within the past few years.

Localization was a different matter: very few believed it at the time, and even fewer saw its importance; among those who failed to fully understand it at first was certainly its author. It has yet to receive adequate mathematical treatment, and one has to resort to the indignity of numerical simulations to settle even the simplest questions about it .

Spin DiffusionFeher, G., Phys. Rev. 114, 1219 (1959); Feher, G. & Gere, E. A., Phys. Rev. 114, 1245 (1959).LightWiersma, D.S., Bartolini, P., Lagendijk, A. & Righini R. “Localization of light in a disordered medium”, Nature 390, 671-673 (1997).Scheffold, F., Lenke, R., Tweer, R. & Maret, G. “Localization or classical diffusion of light”, Nature 398,206-270 (1999).Schwartz, T., Bartal, G., Fishman, S. & Segev, M. “Transport and Anderson localization in disordered two dimensional photonic lattices”. Nature 446, 52-55 (2007).C.M. Aegerter, M.Störzer, S.Fiebig, W. Bührer, and G. Maret : JOSA A, 24, #10, A23, (2007)MicrowaveDalichaouch, R., Armstrong, J.P., Schultz, S.,Platzman, P.M. & McCall, S.L. “Microwave localization by 2-dimensional random scattering”. Nature 354, 53, (1991).Chabanov, A.A., Stoytchev, M. & Genack, A.Z. Statistical signatures of photon localization. Nature 404, 850, (2000).Pradhan, P., Sridar, S, “Correlations due to localization in quantum eigenfunctions od disordered microwave cavities”, PRL 85, (2000)

SoundWeaver, R.L. Anderson localization of ultrasound. Wave Motion 12, 129-142 (1990).

Experiment

Localized StateAnderson Insulator

Extended StateAnderson Metal

f = 3.04 GHz f = 7.33 GHz

Billy et al. “Direct observation of Anderson localizationof matter waves in a controlled disorder”. Nature 453, 891- 894 (2008).

Localization of cold atoms

Roati et al. “Anderson localization of a non-interacting Bose-Einstein condensate“. Nature 453, 895-898 (2008).

Q: What about electrons ?

A: Yes,… but electrons interact with each other

Anderson Model

• Lattice - tight binding model

• Onsite energies εi - random

• Hopping matrix elements Iijj iIij

Iij ={-W < εi <Wuniformly distributed

I < Ic I > IcInsulator All eigenstates are localized

Localization length ξMetal

There appear states extendedall over the whole system

Anderson Transition

I i and j are nearest neighbors

0 otherwise

( ) WdfIc ∗=

Q:Why arbitrary weak hopping I is not sufficient for the existence of the diffusion

?Einstein (1905): Marcovian (no memory)

process g diffusion

j iIij

Quantum mechanics is not marcovian There is memory in quantum propagation!Why?

1ε2ε

Hamiltonian

diagonalize

( ) 221212 IEE +−=− εε

IH diagonalize

( )III

IEE<<−>>−−

≈+−=−12

1212221212 εε

εεεεεε

von Neumann & Wigner “noncrossing rule”Level repulsion

v. Neumann J. & Wigner E. 1929 Phys. Zeit. v.30, p.467

What about the eigenfunctions ?

IH ( )

IEE<<−>>−−

≈+−=−12

1212221212 εε

εεεεεε

What about the eigenfunctions ?

1 1 2 2 1 1 2 2, ; , , ; ,E Eφ ε φ ε ψ ψ⇐

2,12,1

ϕεε

1,22,12,1

ϕϕψ

<<− I

Off-resonanceEigenfunctions are

close to the original on-site wave functions

ResonanceIn both eigenstates the probability is equally

shared between the sites

Anderson insulatorFew isolated resonances

Anderson metalThere are many resonances

and they overlap

Transition: Typically each site is in the resonance with some other one

Condition for Localization:

i j typWε ε= − =

energy mismatch# of n.neighborsI<

energy mismatch

2d=# of nearest neighbors

A bit more precise:

Logarithm is due to the resonances, which are not nearest neighbors

Is it correct?Q:A1:For low dimensions – NO. for

All states are localized. Reason – loop trajectories cI = ∞ 1,2d =

OPhase accumulated when traveling along the loop

The particle can go around the loop in two directions

Memory!

∫= rdp ϕ 21ϕ ϕ=

For d=1,2 all states are localized.

Phase accumulated when traveling along the loop

The particle can go around the loop in two directions

Memory!

∫= rdp ϕ 21ϕ ϕ=

Weak Localization:The localization length can be large

Inelastic processes lead to dephasing, which is characterized by the dephasing length

If , then only small corrections to a conventional metallic behavior

Lϕς >>

Is it correct?Q:A1:For low dimensions – NO. for

All states are localized. Reason – loop trajectories cI = ∞ 1,2d =

A2:Works better for larger dimensions 2d >

A3:Is exact on the Cayley tree

,lncWI K

is the branching number

Anderson Model on a Cayley tree

extended

localized

Eigenfunctions

Does anything interesting happen with the spectrumQ: ?

Density of States

energy

Density of States

energyW− W # I# I−

0I = 0W =

Density of States

energy

0; 0I W≠ ≠

extendedLifshits tail Lifshits tail

Mobility edge Mobility edge

localized and extended never coexist!

DoS DoS

all states arelocalized

I < IcI > Ic

Anderson Transition

- mobility edges (one particle)

extended

Chemicalpotential

Temperature dependence of the conductivity one-electron picture

DoS DoSDoS

( ) 00 >→Tσ ( ) TE Fc

−∝ ( ) TT ∀= 0σ

Assume that all the states

are localized;e.g. d = 1,2 DoS

( ) TT ∀= 0σ

Temperature dependence of the conductivity one-electron picture

Inelastic processestransitions between localized states

β energymismatch

(any mechanism)00 =⇒= σT

Phonon-assisted hopping

Any bath with a continuous spectrum of delocalized excitations down to ω = 0 will give the same exponential

Variable Range HoppingN.F. Mott (1968)

Optimizedphase volume

Mechanism-dependentprefactor

βα εεω −=ω

( ) 00 ==Tσ

Lecture1.2. Phononless conductivity

in Anderson insulators with e-e interaction

1. All one-electron states are localized2. Electrons interact with each other3. The system is closed (no phonons)4. Temperature is low but finite

Given:

DC conductivity σ(T,ω=0)(zero or finite?)

Can hopping conductivity exist without phonons ?

Common belief:

Anderson Insulator weak e-e interactions

Phonon assistedhopping transport

A#1: Sure

Q: Can e-h pairs lead to phonon-less variable range hopping in the same way as phonons do ?

1. Recall phonon-less AC conductivity:Sir N.F. Mott (1970)

2. Fluctuation Dissipation Theorem: there should be Johnson-Nyquist noise

3. Use this noise as a bath instead of phonons

4. Self-consistency (whatever it means)

A#2: No way (L. Fleishman. P.W. Anderson (1980))

A#1: Sure

is contributed by rare resonances

R ∞→

⇒∞→Rmatrix element vanishes

Except maybe Coulomb interaction in 3D

No phonons

No transport T∀

Problem:If the localization length exceeds , then – metal.In a metal e–e interaction leads to a finite

} At high enough temperatures conductivity should be finite even without phonons

A#2: No way (L. Fleishman. P.W. Anderson (1980))

A#1: Sure

A#3: Finite temperature Metal-Insulator Transition(Basko, Aleiner, BA (2006))

insulator

Drudemetal

σ = 0

insulator

metalInteraction strength

Localizationspacing( ) 1−

≡ dνζδζ

Many body localization!

Many body wave functions are localized in functional space

Finite temperature Metal-Insulator Transition

σ = 0

Insulator not

Definitions:0σ =

0d dTσ <Metal

not0σ ≠

0d dTσ >

3. Localization beyond real space

Andrey Kolmogorov

Vladimir Arnold

JurgenMoser

Kolmogorov – Arnold – Moser (KAM) theory

Integrable classical Hamiltonian , d>1:

Separation of variables: d sets of action-angle variables

Quasiperiodic motion: set of the frequencies, which are in general incommensurate. Actions are integrals of motion

1 2, ,.., dω ω ω

2θ2I⊗ ⊗…=>

1 1 1 2 2 2, 2 ; ... , , 2 ;..I t I tθ πω θ πω= =

iI0=∂∂ tIi

A.N. Kolmogorov, Dokl. Akad. Nauk SSSR, 1954. Proc. 1954 Int. Congress of Mathematics, North-Holland, 1957

space Number of dimensionsreal space d

phase space: (x,p) 2denergy shell 2d-1

tori d

Each torus has measure zero on the energy shell !

Integrable dynamics:Each classical trajectory is quasiperiodicand confined to a particular torus, which is determined by a set of the integrals of motion

Andrey Kolmogorov

Vladimir Arnold

JurgenMoser

Kolmogorov – Arnold – Moser (KAM) theoryIntegrable classical Hamiltonian , d>1:Separation of variables: d sets of action-angle variablesQuasiperiodic motion: set of the frequencies,

which are in general incommensurate Actions are integrals of motion

1 2, ,.., dω ω ω

2θ2I⊗ ⊗…=>

;..2,;..,2, 222111 tItI πωθπωθ ==

iI 0=∂∂ tIi

Q:Will an arbitrary weak perturbation of the integrable Hamiltonian

destroy the tori and make the motion ergodic (when each point at the energy shell will be reached sooner or later)

?A: Most of the tori survive

weak and smooth enough perturbations

KAM theorem

A.N. Kolmogorov, Dokl. Akad. Nauk SSSR, 1954. Proc. 1954 Int. Congress of Mathematics, North-Holland, 1957

Andrey Kolmogorov

Vladimir Arnold

JurgenMoser

Kolmogorov – Arnold – Moser (KAM) theory

Q:Will an arbitrary weak perturbation of the integrable Hamiltonian

destroy the tori and make the motion ergodic (i.e. each point at the energy shell would be reached sooner or later)

A:Most of the tori survive weak and smooth enough perturbations

KAM theorem

A.N. Kolmogorov, Dokl. Akad. Nauk SSSR, 1954. Proc. 1954 Int. Congress of Mathematics, North-Holland, 1957 ?

Each point in the space of the integrals of motion corresponds to a torus and vice versa

0ˆ ≠V

Finite motion.Localization in the space of the integrals of motion?

Most of the tori survive weak and smooth enough perturbations

KAM theorem:

2I 0≠

Two integrals of motion

1 1;x yI p I p= =

Rectangular billiard1I

mpLnp ππ

0ˆ ≠V

KAM theorem:

2I 0≠

Energy shell

νµ ,Vνµ

( ) ( ) ( ){ }µµµdIII ,...,1=

Matrix element of the perturbation

One can speak about localization provided that the perturbation is somewhat local in the space of quantum numbers of the original Hamiltonian

( )µµ I

AL hops are local – one can distinguish “near” and “far”KAM perturbation is smooth enough

GlossaryClassical Quantum

Integrable Integrable

KAM LocalizedErgodic – distributed all over the energy shellChaotic

Extended ?

( )IHH

00 = IEH

== ∑ µµµµ

µ ,ˆ0

Strong disorder localizedWeak disorder extended

Strong disorder localized

Moderate disorder extended

No disorder chaotic extended

No disorder integrable localized

Too weak disorder int.localized

Consider an integrable system. Each state is characterized by a set of quantum numbers.

It can be viewed as a point in the space of quantum numbers. The whole set of the states forms a lattice in this space.

A perturbation that violates the integrability provides matrix elements of the hopping between different sites (Anderson model !?)

Is it possible to tell if the states are localized (in some unknown basis) or extended.Q: ?

Density of States is not singular at the Anderson transition

This applies only to the average Density of States !Fluctuations ?

4. Spectral statistics and Localization

Eα - spectrum (set of eigenvalues)- mean level spacing, determines the density of states- ensemble averaging

- spacing between nearest neighbors

- distribution function of nearest neighbors spacing between

Spectral Rigidity

Level repulsion

( ) 4211

,,=∝<<

ββssP

( )sP1

δαα EEs −

ααδ EE −≡ +11

......

RANDOM MATRIX THEORY

N × N N → ∞ensemble of Hermitian matrices with random matrix element

Spectral statistics

Orthogonal β=1

Poisson – completely uncorrelated levels

Wigner-Dyson; GOEPoisson

GaussianOrthogonalEnsemble

Unitaryβ=2

Simplecticβ=4

RANDOM MATRICES

N × N matrices with random matrix elements. N → ∞

Ensembleorthogonalunitary

simplectic

Dyson Ensembles

realizationT-inv potentialbroken T-invariance (e.g., by magnetic field)T-inv, but with spin-orbital coupling

Matrix elementsrealcomplex

2 × 2 matrices

Anderson Model

• Lattice - tight binding model

• Onsite energies εi - random

• Hopping matrix elements Iijj iIij

-W < εi <Wuniformly distributed

Q: What are the spectral statistics of a finite size Anderson model ?

Is there much in common between Random Matrices and Hamiltonians with random potential ?

I < Ic I > IcInsulator

All eigenstates are localizedLocalization length ξ

MetalThere appear states extended

all over the whole system

Anderson Transition

The eigenstates, which are localized at different places

will not repel each other

Any two extended eigenstates repel each other

Poisson spectral statistics Wigner – Dyson spectral statistics

Strong disorder Weak disorder

I0Localized states

InsulatorExtended states

MetalPoisson spectral

statisticsWigner-Dyson

spectral statistics

Anderson Localization andSpectral Statistics

Consider an integrable system. Each state is characterized by a set of quantum numbers.

It can be viewed as a point in the space of quantum numbers. The whole set of the states forms a lattice in this space.

A perturbation that violates the integrability provides matrix elements of the hopping between different sites (Anderson model !?)

Weak enough hopping:Localization - Poisson

Strong hopping:transition to Wigner-Dyson

Extended states:

Level repulsion, anticrossings, Wigner-Dyson spectral statistics

Localized states: Poisson spectral statistics

Invariant (basis independent)

definition

BA, Gefen, Kamenev & Levitov, 1997Basko, Aleiner & BA, 2005. . .

01 1 1

ˆ ˆˆ ˆ ˆ ˆ ˆN N N

z z z x xi i ij i j i i

i i j i iH B J I H Iσ σ σ σ σ

= ≠ = =

= + + ≡ +∑ ∑ ∑ ∑

1,2,..., ; 1i

i N Nσ

Perpendicular fieldRandom Ising model

in a parallel field

- Pauli matrices,

Example: Random Ising model in the perpendicular field

ziσ = ±

Will not discuss today in detail

Without perpendicular field all commute with the Hamiltonian, i.e. they are integrals of motion

01 1 1

= ≠ = =

= + + ≡ +∑ ∑ ∑ ∑

{ }( )0 iH σonsite energy

ˆ ˆ ˆxσ σ σ+ −= +hoping between nearest neighbors

Anderson Model on N-dimensional cube

1,2,..., ; 1i

i N Nσ

Perpendicular fieldRandom Ising model

in a parallel field

- Pauli matrices

{ }ziσ determines a site

Without perpendicular field all commute with the Hamiltonian, i.e. they are integrals of motion

Anderson Model on N-dimensional cubeUsually:# of dimensions

system linear size

d const→

L → ∞

Here:# of dimensions

system linear size

d N= → ∞

01 1 1

= ≠ = =

= + + ≡ +∑ ∑ ∑ ∑

6-dimensional cube 9-dimensional cube

insulator metal

interaction strength

localization spacing( ) 1−

≡ dνζδζMany body localization!

σ = 0

Bad metal

ity σ

temperature TDrude metal

σ > 0

Insulator not

Definitions:0σ =

0d dTσ <Metal

not0σ ≠

0d dTσ >

1D bosons + disorder

Gertsenshtein & Vasil'ev, 1959

1D Localization

Exactly solved: all states are localized

Mott & Twose, 1961Conjectured:. . .

1-particle problemcorrect for bosons as well as for fermions

Bosons without disorder

•Bose - Einstein condensation

•Bose–condensate even at weak enough repulsion

•Even in 1d case at T=0 – “algebraic superfluid”

•Finite temperature – Normal fluid

TNormal fluid

Billy et al. “Direct observation of Anderson localizationof matter waves in a controlled disorder”. Nature 453, 891- 894 (2008).

Localization of cold atoms

Roati et al. “Anderson localization of a non-interacting Bose-Einstein condensate“. Nature 453, 895-898 (2008).

No interaction !

Thermodynamics of ideal Bose-gas in the presence of disorder is a pathological problem: all particles will occupy the localized state with the lowest energy

Need repulsion

1D Bosons + disorder + weak repulsionQ: ?

Weakly interacting bosons

•Bose - Einstein condensation

•Bose–condensate even at weak enough repulsion

•Even in 1D case at T=0 – “algebraic superfluid”

TNormal fluid

2. No disorder1. No interactiondisorder

For any energy at

finite disorder

1D localization

3. Weak repulsiondisorder

Superfluid-insulator transition Superfluid-

insulator transition

T=0 Superfluid – Insulator Quantum Phase Transition

disorder

luid Berezinskii Kosterlitz

Thoulesstransition in 1+1 dim.

T. Giamarchi and H. J. Schulz, Phys. Rev., B37, #1(1988).E. Altman, Y. Kafri, A. Polkovnikov & G. Refael, Phys. Rev. Lett., 100, 170402 (2008).G.M. Falco, T. Nattermann, & V.L. Pokrovsky, Phys. Rev., B80, 104515 (2009).

relatively strong interaction

weakinteraction}

TNormal fluid

disorder

Is it a normal fluid at any temperature?

Dogma There can be no phase transitions at a finite temperature in 1DVan Howe, Landau

Reason Thermal fluctuation destroy any long range correlations in 1D

Neither normal fluids nor glasses (insulators) exhibit long range correlations

True phase transition: singularities in transport (rather than thermodynamic) properties

T=0 Normal fluid – Insulator Phase Transition:

What is insulator?Perfect Insulator

Zero DC conductivity at finite temperatures

Possible if the system is decoupled from any outside bath

Normal metal (fluid)

Finite (even if very small) DC conductivity at finite temperatures

1D Luttinger liquid: bosons = fermions ?Bosons with infinitely strong repulsion ≈ Free fermions

Free bosons ≈ Fermions with infinitely strong attraction

Weakly interactingbosons ≈ Fermions with strong

attractionU

As soon as the occupation numbers become large the analogy with fermions is not too useful

1D Weakly Interacting Bosons + Disorder

disorder

K-Ttransition

“Alg

3. T=0

TNormal fluid

2. No disorder

1. No interactiondisorder

For any temperature

and any finite

disorder1D

localizationdisorder

Aleiner, BA & Shlyapnikov, 2010, Nature Physics, to be published cond-mat 0910.4534

Density of States ν(ε) in one dimension

ν(ε)

No disorder

Quadratic spectrum2

( )επ

εν 222

- singularity

ν(ε) ν(ε)

No disorder

Quadratic spectrum

In the presence of disorder the singularity is smeared

( )επ

εν 222

( )επ

εν 222

ν(ε)

Lifshitz tail: exponentially small Density

of States

Weak disorder – random potential U(x)

ν(ε)

Random potential U(x):Amplitude U0Correlation length σ

Short range disorder:2

0 σmU

Localization length ζ>>σ

Characteristic scales:

ν(ε)

Energy

Length

1 34 20

2E U mσ∗

σςσ∗

≡ >>

( ) 1−∗∗Eς

All states are localizedLocalization length:

ς ες ε ες ε

∗ ∗

∗ ∗∗

ν(ε)

( ) 1−∗∗Eς

ν(ε)

( ) 1−∗∗Eς

Finite density Bose-gas with repulsion

Temperature of quantum degeneracy

nDensity

Two more energy scales

Interaction energy per particle mnTd

ngTwo dimensionless parameters

E ngκ ∗≡ Characterizes the strength of disorder

dng Tγ ≡ Characterizes the interaction strength

Strong disorder 1>>κ1<<γWeak interaction

Dimensionless temperature ngTt =

Critical temperature

( )γκκ ,tcc =cT ( )γκ ,cc tt =

Critical disorder

Phase transition line on the t,κ - plane

1 γ1 γ1

( )1 32c tκ γ=

1 3c tκ =( )c tκ

t T ng≡

E ngκ ∗≡

~ 1cκ

Finite temperature phase transition in 1D

Conventional Anderson Model

Basis: ,i i

i iiH ε0ˆ ∑

ˆnnji

Hamiltonian: 0ˆ ˆ ˆH H V= +

•one particle,•one level per site, •onsite disorder•nearest neighbor hoping

labels sites

Transition: happens when the hoping matrix element exceeds the energy mismatch

The same for many-body localization

• many particles,• several particles

per site.• interaction

Many body Anderson-like ModelBasis:

0,1,2,3,...in =

{ }inµ ≡

occupation numbers

label sitesi

0H Eµµ

µ µ= ∑

Many body Anderson-like ModelBasis: µ

0,1,2,...in =

Hamiltonian:0

ˆ ˆH H V= +

{ }inµ =labels sites

occupation numbers

( )( )

V Iµ η µ

µ η µ= ∑

( ) .., 1,.., 1,.., 1,.., 1,..

, , , . .i j k ln n n n

i j k l n n

δη µ = − − + +

• many particles,• several particles

per site.• interaction

Conventional Anderson

Many body Anderson-like Model

Basis: ilabels sites

i j n n

= +∑

∑( )

H E Iµµ µ ν µ

µ µ µ ν µ= +∑ ∑

Basis: ,µ0,1,2,...in =

{ }inαµ =labels sites

occupation numbersi

“nearest neighbors”:

( ) .., 1,.., 1,.., 1,.., 1,..

, , , . .i j k ln n n n

i j k l n n

δν µ = − − + +

Transition temperature: ( )c cT t ng≡

lk, ,i j k l⇒

transition

Transition temperature: ( )c cT t ng≡

lk, ,i j k l⇒

transition

,ij kl i j k lε ε ε ε∆ ≡ + − − energy mismatch

,ij klI matrix element Decay of a state i

∆ typical mismatch

1N typical # of channelsAnderson condition: I

( ) ( ) ( )( ) ( )

T N TI T

T N T>><<

∆∆

typical matrix element

extendedlocalized

Matrix element of the transition

should be compared with the minimal energy mismatch

High temperatures: dT T>>

Bose-gas is not degenerated; occupation numbers either 0 or 1

( ) 131

>>∝ γκ tttcNumber of channels

( ) ( ) ( )~ ~I g T gE Tς ε ς∗ ∗=

( ) ( ) ( ) 11 2 2 2~n n T Eνς ς ν ς−−

∗ ∗

Localization spacing

1t γ −>>

~I g ς

*,ng Eµ >>

Intermediate temperatures:

dTT <<

1 2 1tγ γ− −<< <<

2. Bose-gas is degenerated; occupation numbers either >>1.

3. Typical energies |µ|=T2/Td , µ is the chemical potential. Correct as long as

4. Characteristic energies

1tγ <<

1t γ >>

ε µ<<

>>We are still dealing with the high energy states

multiple occupation ( ) ~ TN ε

Intermediate temperatures:

dTT <<

Bose-gas is degenerated; typical energies ~ |µ|>>T occupation numbers >>1 matrix elements are enhanced

*2 , EngTT d >>=µ

( ) 13132 <<<<∝ γγγκ tttc

1 2 1tγ γ− −<< <<

( )1 ~ g TINς ε ε

Low temperatures: 1 2t γ −<<

Eµ ∗<<iε µ<

Suppose 1E ngκ ∗≡ >>Bosons occupy only small fraction of low energy states

Start with T=0

Eµ ∗<<

iε µ<

( )i gµ ε ς∗−

( ) ( ) 1Eν ε ς −∗ ∗=

Occupation #:

= Eµ κ∗=

Suppose 1E ngκ ∗≡ >>Bosons occupy only small fraction of low energy states

Localization length

ς∗( )l κ ( )l κ ς κ ς∗ ∗= >>Occupation

( ) 1 2 1nl κ ς γ −∗ = >>

Start with T=0

( )l κ( )l κ ς κ ς∗ ∗= >>

Occupation( ) 1 2 1nl κ ς γ −

∗ = >>

1E ngκ ∗≡ >> “lakes”

( )l κ ς∗>> Strong insulator

Distance

lκ κ

κ ς∗

<<Insulator – Superfluid transition in a chain of “Josephson junctions”

1E ngκ ∗≡ >> Strong insulator

~ 1cκ0T = transition } ~ 1cκ for 1 2t γ −<<

Disordered interacting bosons in two dimensions

Justification:1. At T=0 normal state is unstable with respect to either

insulator or superfluid.2. At finite temperature in the vicinity of the critical

disorder the insulator can be thought of as a collection of “lakes”, which are disconnected from each other. The typical size of such a “lake” diverges. This means that the excitations in the insulator state are localized but the localization length can be arbitrary large. Accordingly the many –body delocalization is unavoidable at an arbitrary low but finite T.

Disordered interacting bosons in two dimensions

Phononless conductance

Many-body Localization of fermions

insulator metal

σ = 0

Bad metal

ity σ

σ > 0

Insulator not

Definitions:0σ =

0d dTσ <Metal

not0σ ≠

0d dTσ >

• many particles,• several levels

per site, • onsite disorder• local

interaction

0H Eµµ

µ µ= ∑

Many body Anderson-like ModelBasis: µ

0,1inα =Hamiltonian:

0 1 2ˆ ˆ ˆ ˆH H V V= + +

{ }inαµ =labels sites

occupation numbers

i labels levelsα

I( ) .., 1,.., 1,.. , , . .i jn n i j n nα βν µ = − + =

( )( )1

µ ν µ

µ ν µ= ∑1V

V Uµ η µ

µ η µ= ∑( ) .., 1,.., 1,.., 1,.., 1,..i i i in n n nα β γ δν µ = − − + +

Conventional Anderson

Many body Anderson-like Model

Basis: ilabels sites

i j n n

= +∑

( ) .., 1,.., 1,.. , , . .i jn n i j n nα βν µ = − + =

( )( )

µ ν µ

µ η µ

µ ν µ

µ η µ

( ) .., 1,.., 1,.., 1,.., 1,..i i i in n n nα β γ δη µ = − − + +

Basis: ,µ0,1inα =

{ }inαµ =

labels sites occupation

numbersi labels

levelsαi

Two types of “nearest neighbors”:

2. Add an infinitesimal Im part iη to Eµ

4 1)2) 0

→ ∞→limits

insulator

1. take descrete spectrum Eµ of H0

3. Evaluate ImΣ µ

Anderson’s recipe:

4. take limit but only after ∞→N5. “What we really need to know is the

probability distribution of ImΣ, not its average…” !

0→η

Probability Distribution of Γ=Im Σ

insulator

Look for:

η is an infinitesimal width (Impart of the self-energy due to a coupling with a bath) of one-electron eigenstates

Stability of the insulating phase:NO spontaneous generation of broadening

0)( =Γ εαis always a solution

ηεε i+→linear stability analysis

222 )()(

)( αα

α ξεξεπδ

ξε −Γ

+−→Γ+−

After n iterations of the equations of the Self Consistent Born Approximation

nTconstP

Γ∝Γ

λδλη

1ln)( 23

firstthen (…) < 1 – insulator is stable !

Physics of the transition: cascades

Conventional wisdom:For phonon assisted hopping one phonon – one electron hop

It is maybe correct at low temperatures, but the higher the temperature the easier it becomes to create e-h pairs.

Therefore with increasing the temperature the typical number of pairs created nc (i.e. the number of hops) increases. Thus phonons create cascades of hops.

Typical size of the cascade

Localization length≈

Physics of the transition: cascades

Conventional wisdom:For phonon assisted hopping one phonon – one electron hop

It is maybe correct at low temperatures, but the higher the temperature the easier it becomes to create e-h pairs.

Therefore with increasing the temperature the typical number of pairs created nc (i.e. the number of hops) increases. Thus phonons create cascades of hops.

At some temperature This is the critical temperature. Above one phonon creates infinitely many pairs, i.e., phonons are not needed for charge transport.

( ) .∞→= TnTT cc

transition !mobility edge

Many-body mobility edge

Large E (high T): extended states

Fermi Golden Rule hopping (bad metal)

transition !mobility edge

(good metal)

Many-body mobility edge

Finite T normal metal – insulator transition is another

example of the many-body localization

insulator metal

σ = 0

Bad metal

ity σ

σ > 0

nonergodic ergodic

Definition: We will call a quantum state ergodic if it occupies the number of sites on the Anderson lattice, which is proportional to the total number of sites :

0 → ∞→NNNµ

nonergodic

0> → ∞→ constNN

ergodic

Localized states are obviously not ergodic: constN N → ∞→µ

Is each of the extended state ergodicQ: ?A: In 3D probably YES, for d>4 - probably NO

Cayley tree(Bethe lattice)

Nonergodic states

is the branching number

cI I W< <Extended but not ergodic

lnWI N N NK µ≈ ⇒ ≈ <<

nonergodic

glassy???

nonergodic

Main postulate of the Gibbs StatMech-equipartition (microcanonical distribution): In the equilibrium all states with the same energy are realized with the same probability.Without interaction between particles the equilibrium would never be reached – each one-particle energy is conserved.Common believe: Even weak interaction should drive the system to the equilibrium.

Is it always true?

Lecture 3. 4. Speculations

insulator metal

σ = 0

Bad metal

ity σ

σ > 0

Q: ?What happens in the classical limit 0→

Speculations: 1.No transition2.Bad metal still exists

0cT →

Reason: Arnold diffusion

Arnold diffusion

Each point in the space of the integrals of motion corresponds to a torus and vice versa

0ˆ ≠V

Finite motion?

2d = All classical trajectories correspond to a finite motion

2d > Most of the trajectories correspond to a finite motion

However small fraction of the trajectories goes infinitely far

Arnold diffusion

space # of dimensionsreal space dphase space 2denergy shell 2d-1

tori d

1. Most of the tori survive – KAM

2. Classical trajectories do not cross each other

.2 1en shell torid d d= ⇒ − =

Each torus has “inside” and “outside” inside

.2 1en shell torid d d= ⇒ − =

A torus does not have “inside” and “outside” as a ring in >2 dimensions

Speculations:1. Arnold diffusion Nonergodic (bad)

2. Appearance of the transition (finite Tc ) –quantum localization of the Arnold diffusion

ConclusionsAnderson Localization provides a relevant language for description of a wide class of physical phenomena – far beyond conventional Metal to Insulator transitions.Transition between integrability and chaos in quantum systemsInteracting quantum particles + strong disorder.Three types of behavior:

ordinary ergodic metal“bad” nonergodic metal“true” insulator

A closed system without a bath can relaxation to a microcanonical distribution only if it is an ergodic metal

Boris Altshuler - Lancaster University · Introduction to Anderson Localization 2. Phononless...

Documents

Boris Red 4.2 Release Notes - Boris FX

FromOpen Access toOpen Science ... · > Introduction > BORIS Publications —UniBelook-up —ORCID implementation > BORIS Theses > BORIS CRIS > BORIS Research Data ... Advanced ORCID

Boris Gleb

Disorder and chaos in quantum system: Anderson localization and its generalization Boris Altshuler (Columbia) Igor Aleiner (Columbia) (6 lectures)

Disorder and chaos in quantum systems II. Lecture 2. Boris Altshuler Physics Department, Columbia University

Disorder and chaos in quantum systems II. Lecture 3. Boris Altshuler Physics Department, Columbia University

Boris Bells

math.haifa.ac.ilmath.haifa.ac.il/chorny/publications/Chorny - LocalizationI.pdf · LOCALIZATION WITH RESPECT TO A CLASS OF MAPS I {EQUIVARIANT LOCALIZATION OF DIAGRAMS OF SPACES BORIS

Many Body Anderson Localization - UCSBonline.kitp.ucsb.edu/online/electroglass10/altshuler/pdf/...No disorder disorder glass (ator) For any energy at finite disorder 1D localization

Evil Boris

I.L. Aleiner ( Columbia U, NYC, USA ) B.L. Altshuler ( Columbia U, NYC, USA ) K.B. Efetov ( Ruhr-Universitaet,Bochum, Germany) Localization and Critical

I.L. Aleiner ( Columbia U, NYC, USA ) B.L. Altshuler ( Columbia U, NYC, USA )

Boris Altshuler Columbia University Anderson Localization against Adiabatic Quantum Computation Hari Krovi, Jérémie Roland NEC Laboratories America

FIRM RESUME June 2016 Altshuler Berzon LLP is a San Francisco

Boris Spassky

Boris Red Tutorial - Boris Red, web reference

Boris Yeltsin

Revista Boris

Jill Altshuler Reservoir High School Student Services NAVIGATING THE COLLEGE SEARCH PROCESS

David Reichman Laura Kaufman Mike Steigerwald Stephen O’Brien Ruben Gonzalez Mark Hybertsen Boris Altshuler Igor Aleiner